Currently, targeted agents are in clinical testing for treatment of ER- breast cancer, but have not been adequately tested for prevention. This is because, without a better understanding of the biology of human breast cancer initiation, Phase I/II testing of targeted agents for preventing ER- breast cancer will be too risky and expensive. To meet these challenges, we aim to identify key signaling networks activated during initiation of ER- breast cancer in women in our high-risk cohort and use these networks to target prevention. Normal mammary gland homeostasis requires the coordinated regulation of signaling networks. Currently we lack an understanding of 1) whether the signaling networks that are activated in aggressive ER- breast cancer are also activated in mammary atypia, and 2) if so, whether activation of specific signaling networks predicts cancer initiation and progression. Here we will investigate whether signaling network activation in atypia predicts subsequent cancer etiology. Activation of Akt/mTOR, IL6/Stat3, EGRF/MEK/ERK, and mitochondrial survival pathways are known to predict aggressive biology in ER- breast cancer. Our Preliminary Data provide evidence that activation of these poor-prognosis signaling networks can be detected in mammary atypia from high risk women. However, just because we can identify signatures with high inter-class coefficients of variation between subjects, does not necessarily imply those processes are important for cancer etiology. Here will test the hypothesis that signaling pathways that underlie the aggressive behavior of ER- breast cancer can be detected in mammary atypia from high-risk women, and whether these signaling pathways can be used to predict cancer initiation and guide targeted prevention strategies. Aim 1 will test whether phosphoprotein signatures identified in ER- breast cancers are present in premalignant lesions in high-high risk women. Aim 2 will prospectively investigate whether phosphoprotein signatures predict initiation of ER- breast cancer. Aim 3 will investigate whether RPPM signatures predict in vitro sensitivity to targeted agents. Aim 4 will perform Pilot Testing and use phosphoprotein signatures to select and track response to targeted agents to prevent breast cancer in high-risk women with mammary atypia. Significance: Here we will identify protein signaling pathways that are activated during breast cancer initiation. Information gained in this proposal will allow us to identify activated signaling pathways in atypia, test whether pathway activation predicts cancer etiology, and use this information to select, and track response to, targeted agents.